INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rose

-0.75

tnc

-0.71

acea

-0.67

 greeted

-0.65

ãĤ´ãĥ³

-0.65

acy

-0.64

,,,,

-0.64

\":

-0.63

Mom

-0.63

 resil

-0.62

POSITIVE LOGITS

unes

0.75

ements

0.73

BLIC

0.69

hap

0.67

voy

0.64

focus

0.63

incial

0.62

ADVERTISEMENT

0.60

inion

0.59

 Tong

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.